Vol. 510: 241–253, 2014 MARINE ECOLOGY PROGRESS SERIES Published September 9 doi: 10.3354/meps10798 Mar Ecol Prog Ser

Contribution to the Theme Section ‘ blooms and ecological interactions’

Asexual reproduction strategies and blooming potential in

Agustín Schiariti1,2,*, André C. Morandini3, Gerhard Jarms4, Renato von Glehn Paes3, Sebastian Franke4, Hermes Mianzan1,2

1Instituto Nacional de Investigación y Desarrollo Pesquero (INIDEP), Paseo V, Ocampo No. 1, B7602HSA Mar del Plata, Argentina 2Instituto de Investigaciones Marinas y Costeras (IIMyC), CONICET, Universidad Nacional de Mar del Plata, 7600 Argentina 3Departamento de Zoologia, Instituto de Biociências, Universidade de São Paulo (USP), Rua do Matão trav. 14 n. 101, São Paulo, 05508-090 SP, Brazil 4Biocenter Grindel and Zoological Museum, University of Hamburg, Martin-Luther-King Platz 3, 20146 Hamburg, Germany

ABSTRACT: Scyphistomae show different modes of propagation, occasionally allowing the sudden release of great numbers of medusae through strobilation leading to so-called jellyfish blooms. Accordingly, factors regulating asexual reproduction strategies will control scyphistoma density, which, in turn, may influence blooming potential. We studied 11 scyphistoma species in 6 combinations of temperature and food supply to test the effects of these factors on asexual repro- duction strategies and reproduction rates. Temperature and food availability increased reproduc- tion rates for all species and observed reproduction modes. In all cases, starvation was the most important factor constraining the asexual reproduction of scyphistomae. Differences in scyphis- toma density were found according to the reproductive strategy adopted by each species. Differ- ent Aurelia lineages and Sanderia malayensis presented a multi-mode strategy, developing up to 5 propagation modes. These species reached the highest densities, mostly through lateral budding and stolons. Cassiopea sp., cephea, Mastigias papua and Phyllorhiza punctata adopted a mono-mode reproductive strategy, developing only free-swimming buds. Lychnorhiza lucerna, Rhizostoma pulmo and Rhopilema esculentum also presented a mono-mode strategy, but they only developed podocysts. These 3 species had the lowest reproduction rates and densities; not only their reproduction rates but also the need for a 2-fold set of environmental stimuli to pro- duce new polyps (one for encystment, another for excystment) made this reproduction mode the slowest of those observed to be utilized for propagation. We conclude that blooms may be defined phylogenetically by the specific asexual modes each species develops, which, in turn, is regulated by environmental conditions.

KEY WORDS: Polyp · Scyphistomae · Mono-mode strategy · Multi-mode strategy · Budding · Podocysts

Resale or republication not permitted without written consent of the publisher

INTRODUCTION strobilation), leading to population pulses that can sometimes result in medu sa outbreaks or the so- A main feature of the majority of blooming called jellyfish blooms (CIESM 2001). Accordingly, scyphozoans is the presence of a benthic polyp it is logical to assume that once strobilation is trig- stage during which several asexual reproduction gered, the greater the number of scyphistomae, the modes are utilized for propagation. In turn, under higher the number of released medusae. Therefore, specific environmental conditions polyps produce factors regulating polyp density will influence the and release great numbers of ephyrae (through formation of medusae blooms.

*Corresponding author: [email protected] © Inter-Research 2014 · www.int-res.com 242 Mar Ecol Prog Ser 510: 241–253, 2014

Different environmental factors have been pro- ent asexual reproduction modes have been described posed to influence scyphistoma asexual reproduction for several species, little is known about the impor- and mortality rates. Among them, temperature and tance of each of them in terms of polyp propagation. food availability appear to be the most important To address these questions we performed a series ones for the majority of species (Lucas et al. 2012). In of experiments involving 11 scyphistoma species. general, reproduction rates increase with tempera- These experiments were organized to evaluate the ture and food supply. However, the results available effect of different combinations of temperature and in the literature are conflicting and have hindered food supply on the asexual reproductive strategy of our understanding of reproductive patterns, as made each species. The null hypotheses tested were that evident in the review by Lucas et al. (2012). reproductive strategy, reproduction rates and polyp One of the reasons behind these controversial density did not differ according to species or experi- results may be found in the diversity of asexual mental conditions. reproductive strategies existing among scyphozoans and their interactions with multiple environmental factors. Only a handful of studies have discriminated MATERIALS AND METHODS between the different asexual modes scyphistomae exhibit during propagation (e.g. Han & Uye 2010), Scyphistomae were reared in laboratory cultures with the majority of studies focusing on the formation from specimens collected in their natural environ- of podocysts and ‘buds’. However, non-coronate ment or obtained from artificial fertilization. All spe- scyphozoans multiply through a variety of asexual cies were maintained inside incubators under the reproduction modes (Adler & Jarms 2009). Although temperatures and salinities provided in Table 1. the final product of these propagation modes is in all Scyphistomae were kept in the dark and fed once cases a new polyp, the use of a particular mode, or a weekly with newly hatched Artemia nauplii. Cul- combination of more than one mode, might have eco- tures were maintained in glass or polystyrene dishes logical implications related to blooming potential. (~200 ml) filled with filtered (5 µm) seawater (34 psu). While some species exhibit the potential to display Although Cassiopea sp., Cephea cephea, Mastigias several asexual modes, even simultaneously, others papua and Phyllorhiza punctata polyps usually host only exhibit a few or just one (Adler & Jarms 2009). zooxanthellae in their tissues (which are important Therefore, considering the reproductive strategy to for strobilation), no studies have indicated that dark- be the number and type of asexual reproduction ness influences other types of asexual reproduction modes displayed and the relative importance of each (except for that of Cassiopea andromeda; Hofmann et mode to polyp density, we hypothesize that those al. 1978). species capable of switching their reproductive stra- Combined effects of temperature and food supply tegy in response to environmental fluctuations will were observed separately for 11 scyphozoan species. have a greater blooming potential. Although differ- The experiments included 3 different temperatures

Table 1. List of species, and culture and experimental conditions used in the present study. T2: culture conditions prior to the experiment; T1 and T3: T2 − 5ºC and T2 + 5ºC, respectively. HL: Helgoland; JP: Japan; FR: France

Species Track culture/origin Culture conditions Experimental conditions Temp. Salinity Temp. (ºC): Salinity (ºC) T1, T2, T3

Aurelia sp.HL Helgoland, Germany 10 34 5, 10, 15 34 Aurelia sp.JP Kagoshima Bay, Japan 20 34 15, 20, 25 34 Aurelia sp.FR Roskoff, France 20 34 15, 20, 25 34 Sanderia malayensis Same as in Adler & Jarms (2009) 20 34 15, 20, 25 34 Cassiopea sp. Cabo Frio, Brazil 20 34 15, 20, 25 34 Cephea cephea Berlin Zoo Aquarium 20 34 15, 20, 25 34 Mastigias papua Berlin Zoo Aquarium 20 34 15, 20, 25 34 Phyllorhiza punctata Berlin Zoo Aquarium 20 34 15, 20, 25 34 Lychnorhiza lucerna San Clemente del Tuyú, Argentina 20 22 15, 20, 25 22 Rhizostoma pulmo Mar Menor, Spain 15 34 10, 15, 20 34 Rhopilema esculentum Kamo Aquarium, Japan 15 34 10, 15, 20 34 Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 243

(maintained by incubators; Table 1) and 2 feeding ied factors because interaction could not be deter- conditions (starvation, feeding ad libitum). Fully mined due to missing data (no polyps or no reproduc- developed scyphistomae showing no external evi- tive products). In these cases, we used a t-test, 1-way dence of reproduction were detached from the stem ANOVA, or the non-parametric Mann-Whitney and cultures and placed in experimental units (glass Kruskal-Wallis tests depending on whether or not dishes containing 150 ml of filtered seawater). Scyp - each data set met the as sumptions. histomae were allowed to reattach and acclimate to the experimental temperatures for 1 wk without food. Dishes were covered by glass plates to avoid eva - RESULTS poration. They were held in the dark, except during observation periods (~30 min observation−1) and Asexual reproduction modes water exchange. Each experimental unit included 1 individual scyphistoma. Three replicates were uti- We observed the following asexual reproduction lized for each combination of temperature and food modes: typical lateral budding (LB), lateral budding supply, accounting for a total of 18 initial polyps per by means of stolons (LBst), ‘Sanderia-type’ budding species. Experimental temperatures were chosen (StyB), reproduction from parts of stolons/stalks (ST), based on species culture conditions (Table 1). typical free-swimming buds/planuloids (FSB), motile Scyphistomae were fed twice weekly with newly bud-like tissue particles (MP) and podocysts (POD) hatched Artemia nauplii. After the scyphistomae fed (Table 2, Fig. 1). for 2 h, seawater, debris and uneaten food were dis- Three different asexual reproduction modes in - carded and replaced with filtered seawater at the volved the formation of stolons. In the LBst, a bud corresponding temperature and salinity conditions. emerges from a stolon giving place to a new polyp The presence of uneaten food items in all cases after (Fig. 1B) (see description in Adler & Jarms 2009). On the feeding period guaranteed that ad libitum condi- the other hand, once a stolon is formed, this new tions had been provided. stolon (or the old stalk) is constricted and severed, Prior to each feeding event, the following variables leaving variable amounts of tissue attached near the were registered: (1) specific asexual reproduction mother polyp from where new scyphistomae deve - modes, (2) the number of polyps produced during the lop. These tissue fragments were registered as re - 6 wk experimental period, i.e. polyp density (PD), (3) productive products and designated ST (Fig. 1D). the total number of reproductive products produced Finally, formation of PODs occurred beneath the during the 6 wk experimental period, i.e. overall reproduction rates (Rr) and (4) the Table 2. Observed asexual reproduction modes for the studied species number of each type of reproductive prod- (for details on the reproductive modes see the ‘Results’ section). LB: uct produced during the 6 wk ex perimental typical lateral budding; LBst: lateral budding by means of stolons; StyB: period, i.e. specific reproduction rates. The Sanderia-type budding; ST: reproduction from parts of stolons/ stalks; different asexual reproduction modes were FSB: typical free-swimming buds/planuloids; MP: motile bud-like tissue particles; POD: podocysts; HL: Helgoland; JP: Japan; FR: France identified according to Adler & Jarms (2009). All reproductive products unable to Species LB LBst StyB ST FSB MP POD feed (motile or non-motile, attached to or detached from the mother polyp) were con- MONO-MODE sidered for Variables 3 and 4. Once tentacles FSB-producers had developed and feeding had started, Cassiopea sp. −−−− −−  they were considered new individuals and Cephea cephea −−−− −− Mastigias papua −−−− −− considered in PD measurements. Phyllorhiza punctata −−−− −− Null hypotheses were examined by 2-way POD-producers ANOVA or Kruskal-Wallis tests depending Lychnorhiza lucerna −−−−−− on whether each data set met the assump- Rhizostoma pulmo −−−−−−  tions of normality and homocedasticity. In Rhopilema esculentum −−−−−− case of significant differences either Bonfer- MULTI-MODE   roni’s or Dunn’s post hoc test was per- AureliaHL − −−−    formed. In some cases of very low or null AureliaFR − − AureliaJP −  −  reproduction, or mortality of scyphistomae, Sanderia malayensis −−−  − we chose not to test the data for the 2 stud- 244 Mar Ecol Prog Ser 510: 241–253, 2014

Fig. 1. Different asexual reproduction modes within Scyphozoa. (A) Typical lateral budding (LB; Aurelia sp.); (B) lateral bud- ding by means of stolons (LBst; Aurelia sp.); (C) ‘Sanderia-type’ budding (StyB; Sanderia malayensis); (D) reproduction from parts of stolons/stalks (Aurelia sp.), successive increasing of tension in the stolon with arrows showing disruption points (1−2) and club-shaped tissue particle remained after disruption (3); (E,F) typical free-swimming buds/planuloids (FSB; E: Cassiopea sp., F: Cephea cephea); (G−J) motile bud-like tissue particles (MP; G: formation of the bud in S. malayensis; H, I, J: develop- ment of the motile bud after being released, Aurelia sp.); (K−M) podocysts (POD; K: Aurelia sp., L,M: Lychnorhiza lucerna). Scale bars = 100 µm (A, B, C, K, L, M); 200 µm (G, H, I); 500 µm (D, E, F, J) Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 245

basal region of the scyphistomae or at the attachment during experiments. The first main group was repre- area of stolons (or alternatively both). Independent of sented by species displaying only 1 type of repro - its origin, POD formation always involved the devel- duction, i.e. the mono-mode species. Among these opment of stolons (Fig. 1K,L). These stolons, although we defined 2 subgroups: one comprised species pro- identical to other stolons, were not counted as the ducing only FSBs (Cassiopea sp., Cephea cephea, mode ST. Masti gias papua and Phyllorhiza punctata) and the Survival was high, reaching nearly 100% in all other included species that only reproduce through treatment combinations for all species, indicating an PODs (Lychnorhiza lucerna, Rhizostoma pul mo and overall tolerance to experimental conditions. The Rhopilema esculentum) (Table 2, Fig. 1). The second number of asexual reproduction modes displayed by main group of species, i.e. the multi-mode species, each species varied between 1 and 5 (Table 2). We comprised Sanderia malayensis and the 3 Aurelia divided the species into 2 main groups according to strains. These species reproduced through up to 5 dif- the numbers of reproductive modes they exhibited ferent asexual reproduction modes (Table 2, Fig. 1).

Overall effects of temperature and food supply on asexual reproduction

Asexual reproduction rates in crea- sed with food availability and warmer temperatures for all species and observed reproduction modes (see Figs. 2−7).

Effects on mono-mode group

FSB-producers

In general, Rr increased with food supply and temperature (Fig. 2). No re- production was observed under starva- tion conditions at 15ºC. Highest bud- ding rates were reached by fed polyps at the highest temperatures (Cassiopea sp. and C. cephea: 25ºC; M. papua and P. punctata: 20 and 25ºC) (Fig. 2). Un- derlying this basic pattern, the 4 spe- cies presented different responses to experimental conditions. No budding was observed under starvation for Cas- siopea sp. or C. cephea (Fig. 2). Con- versely, M. papua and P. punctata pro- duced buds during starvation, although at lower rates compared to fed polyps (Fig. 2). In M. papua, budding was ob- served only at 20 and 25ºC. This species was not able to reproduce at 15ºC re gardless of food availability. In Fig. 2. Effect of temperature and food supply on free-swimming bud/planuloid contrast, upon reaching a critical tem- production (FSB) and polyp density (PD, number of polyps in each experimen- tal unit) of different Scyphozoa after 6 wk. Data are mean numbers of each perature of 20ºC, this species started variable ± standard deviation. Different letters represent significant differ- budding at rates that were elevated by ences at α = 0.05 food availability, but not affected by 246 Mar Ecol Prog Ser 510: 241–253, 2014

higher temperatures (Fig. 2). P. punctata showed the AureliaFR lowest budding rates of this group. However, P. punc- tata produced a higher number of buds than the other Rr increased with temperature and food supply species at 15ºC and when food was available (Fig. 2). (2-way ANOVA; p < 0.01), and the effects of the interaction between the 2 factors were not significant (p > 0.05) (Fig. 4A). Rr increased with temperature POD-producers in fed polyps (Bonferroni’s post hoc test; p < 0.01) (Fig. 4A). The effects on starved polyps were not sig- The highest number of PODs was produced under nificant (Bonferroni’s post hoc test; p > 0.05) (Fig. 4A). warmer temperatures while feeding ad libitum PD exhibited a similar but sharper response to the (Fig. 3). Excystment was observed only for L. lu cer na. studied factors (Fig. 4B).

No significant effects of the treatments were ob - Forty-eight percent of the new AureliaFR scy- served on excystment rates (Fig. 3). phistomae were produced through LB; 32%, by ST; 17%, by LBst; and 3%, by MPs. The effects of temperature and food supply on the specific Effects on multi-mode group asexual modes follow ed the same pattern described for Rr (Fig. 4C−H). LB and LBst rates were very The Aurelia strains and S. malayensis underwent low, or even null under starvation, and were various asexual reproduction modes, including LB, enhanced by warmer temperatures (Bonferroni’s LBst, StyB, ST, MP and POD (Table 2, Fig. 1). With post hoc test; p < 0.01) (Fig. 4C−F). The ST rate the exception of the Cassiopea sp. FSB rate at 25ºC, followed a similar pattern (Mann-Whitney test; p < the multi-mode species presented higher reproduc- 0.05) (Fig. 4G,H). The few MPs observed (n = 6) tion rates than the mono-mode species. S. malayensis were produced only by fed polyps at 15ºC. Four reached the highest production rates (383 new po - PODs were produced by this species, but no lyps in 6 wk). excystment was observed.

AureliaJP

The effects of temperature and food availability on the Rr of this species (Fig. 5A) followed the same pattern

described for AureliaFR (2-way ANO - VA; p < 0.01). Nearly no polyps were developed under starvation (Fig. 5B).

Fifty percent of the new AureliaJP scyphistomae were produced by means of LB; 27%, by ST; 18%, by LBst; and 5%, by MPs. Temperature and food supply effects on the specific asexual modes followed the same pattern described for Rr (Fig. 5C−H). No LBst were produced by starved polyps or fed polyps at the lowest temperature (15ºC). The LBst rate was enhanced by warmer tempera- Fig. 3. Effect of temperature and food tures (Mann-Whitney test; p < 0.05) supply on podocyst production (POD) and polyp density (PD, number of (Fig. 5E). LBst developed into new polyps in each experimental unit) of scyphistomae only at 25ºC (Fig. 5F). different Scyphozoa after 6 wk. Data The general pattern observed for ST are mean numbers of each variable ± was the same (Fig. 5G,H) (Mann- standard deviation. Different letters represent significant differences at Whitney test; p < 0.05). The few MPs α = 0.05 observed (n = 4) were produced by Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 247

only fed polyps at 25ºC. All MPs developed into new when food was supplied (Bonferroni’s post hoc test; scyphistomae. No PODs were produced. feeding: p < 0.01; starving: p < 0.05) (Fig. 6A). PD under ad libitum feeding displayed the same pattern described for Rr (Bonferroni’s post hoc test; p < 0.05)

AureliaHL (Fig. 6B). No effect of temperature on PD was ob - served under starvation (Bonferroni’s post hoc test; The effects of temperature and food availability on p > 0.05) (Fig. 6B). Fifty-five percent of the new Aure-

Rr followed the pattern described for the other liaHL scyphistomae were produced by means of LB; Aurelia strains (2-way ANOVA; p < 0.01) (Fig. 6A), 25%, by LBst; and 20%, by ST. Neither MPs nor but interaction effects were significant (p < 0.05) PODs were observed. The effects of temperature and (Fig. 6A). Rr increased with temperature under both food supply on the specific asexual modes displayed feeding conditions, although reaching higher rates by this species followed the same pattern previously described (Fig. 5C−H).

Sanderia malayensis

The effects of temperature and food availability on Rr followed the gen- eral pattern previously described for the Aurelia strains (2-way ANOVA; p < 0.01) (Fig. 7A). The few scyphis- tomae produced under starvation ap - peared during the first 2 wk of the ex periment. Rr and PD were in- creased by temperature under ad libi- tum feeding (Bonferroni’s post hoc test; p < 0.01), but their effects were not significant under starvation (Bonferroni’s post hoc test; p > 0.05) (Fig. 7A,B). Although reaching differ- ent specific rates, the 3 asexual modes utilized by S. malayensis fol - lowed the same pattern as that de - scribed for Rr (Fig. 7C−H).

DISCUSSION

Our results showed that overall re- production rates and polyp density increase with higher temperatures and food supply for all species. This suggests that spring to summer is the period in which scyphistomae in- crease their population size and in which plankton generally shows its annual peak. However, this trend can Fig. 4. Effects of temperature and food supply on reproduction rates and polyp differ when specific modes of repro- density (PD, number of polyps in each experimental unit) of AureliaFR scyphis- duction are considered, such as PODs tomae after 6 wk. Rr: overall reproduction rates; LB: lateral budding; LBst: lat- in Aurelia (see Han & Uye 2010, eral budding by means of stolon; ST: reproduction from parts of stolons/stalks. Data are mean numbers of each variable ± standard deviation. Different Thein et al. 2012). In general, repro- letters represent significant differences at α = 0.05 duction was very low when no food 248 Mar Ecol Prog Ser 510: 241–253, 2014

of temperature and food supply may synergistically affect scy phisto ma re- production and, subsequently, densi- ties in a positive or negative way. Although 8 different reproductive modes have been described for Scy - pho zoa, only Aurelia species and Sanderia malayensis fit into the multi- mode group. Several examples of other species exhibiting more than 1 mode have been described, such as LB in Catostylus mosaicus, Rhizo- stoma octopus and R. pulmo (Paspal- eff 1938, Pitt 2000, Holst et al. 2007, Straehler-Pohl 2009, Fuentes et al. 2011), LBst and longitudinal fission in R. pulmo (S. Franke pers. obs.) and Cassiopea andromeda (Gohar & Eisawy 1960, Hofmann et al. 1978) and MP in C. cephea (A. Schiariti pers. obs.). However, these observa- tions must be considered exceptional cases with negligible contributions to polyp density. The species displaying multi-mode strategies possessed the highest re- pro duction rates. Aurelia and S. ma - lay ensis achieved substantially high - er reproduction rates than the species reproducing by mono-mode stra - tegies. Among the latter species, the FSB-producers presented higher rates than the POD-producers. Even though lacking a description of spe- cific reproduction modes, Purcell et Fig. 5. Effects of temperature and food supply on reproduction rates and polyp al. (2012) observed the same relation- density (PD, number of polyps in each experimental unit) of AureliaJP scyphis- ship between reproduction rates and tomae after 6 wk. Rr: overall reproduction rates; LB: lateral budding; LBst: lat- eral budding by means of stolon; ST: reproduction from parts of stolons/stalks. reproductive strategies. These authors Data are mean numbers of each variable ± standard deviation. Different reported the highest ‘budding’ rates letters represent significant differences at α = 0.05 reached for the multi-mode A. aurita, the lowest for the POD- producer R. was available, for all species and temperatures. Star- pulmo, with intermediate values reached by the FSB- vation appeared to be a common limiting factor for producer Cotylorhiza tuberculata (see Kikinger 1992 reproduction under all temperature values consid- for reproduction strategy of C. tuberculata). Thein et ered in this study. On the other hand, temperature al. (2013) also reported higher reproduction rates of also affected reproduction rates, but significant ef- Aurelia compared to the POD-producers Chrysaora fects varied between species. Beyond the reduced pacifica and Cyanea nozakii. metabolic rates, the negative effect of temperature Yet, what matters about reproduction rates is not on scyphistoma reproduction was related to the dete- the number of available modes a species can display rioration of tentacles which reduced feeding capabil- but how rapidly these modes can develop functional ity, as we observed for Cassiopea sp. and Cephea polyps. Aurelia and S. malayensis can display most of cephea at the lowest temperature (15ºC) and for Au- the asexual reproduction modes described for the relia at 5 and 10ºC. Therefore, the combined effects Scyphozoa (Adler & Jarms 2009, present study). Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 249

allowing increased energy storage for reproductive and somatic growth. The same pattern has been de scribed for A. aurita s.l. by Han & Uye (2010). These ‘speedy’ reproduction modes are present in multi-mode species such as A. aurita, A. limbata, A. labiata and S. malayensis (Adler & Jarms 2009 and references therein, present study). Some of these modes have also been described for other species like Phacellophora camtschatica, Chry sa - ora hyso scella, C. quinquecirrha, Cy - a nea capillata, C. lamar kii, Catostylus mosaicus, R. pulmo and R. octopus (Lambert 1935, Gröndahl & Hernroth 1987, Pitt 2000, Widmer 2006, Holst et al. 2007, Adler & Jarms 2009 and ref- erences therein, Straehler-Pohl 2009, Fuentes et al. 2011, Holst 2012). How- ever, as stated above, these reports describe exceptional cases that con- tribute very little in terms of the popu- lation’s reproductive output. In addition to their high Rr, the formation of polyps by means of LB, LBst, or StyB produces scyphistomae right beside the mother polyp, allow- ing rapid colonization of available substrates when conditions are opti- mal. On the other hand, Han & Uye (2010) and Thein et al. (2012) ob- served that A. aurita encyst only under conditions of starvation or food scarcity. Besides, Aurelia start devel- oping MPs once a certain threshold in Fig. 6. Effects of temperature and food supply on reproduction rates and polyp density is achieved (the carrying density (PD, number of polyps in each experimental unit) of AureliaHL scyphis- tomae after 6 wk. Rr: overall reproduction rates; LB: lateral budding; LBst: lat- capacity of the population) (Melica eral budding by means of stolon; ST: reproduction from parts of stolons/stalks. 2013). In turn, Aurelia scyphistomae Data are mean numbers of each variable ± standard deviation. Different can detach from the substrate and α letters represent significant differences at = 0.05 drift with currents in response to over- crowding (Melica 2013). Thus, the However, they propagate almost exclusively (95 to multi-mode reproduction strategy allows Aurelia to 100%) through LB (or StyB), LBst and ST when food colonize the available substrate rapidly when condi- is available. The utilization of these modes (mostly tions are favourable, to encyst and withstand starva- LB) allows a very rapid colonization of available tion and other adverse conditions and to develop −1 substra tes (AureliaFR: up to 2.2 scyphistomae d ; S. motile buds or detach in order to facilitate dispersion malayensis: up to 4.5 scyphistomae d−1). The forma- and avoid the negative effects of inter- or intraspe - tion of a new scyphistoma takes between 3 and 5 d cific competition for space and food. The potential from the first signs of the bud to detachment of the capability to switch its reproductive strategy in new scyphistoma. The new bud develops its tenta- response to environmental clues gives Aurelia high cles and starts feeding before detachment, becoming adaptability. Considering these features, in addition a double-headed polyp with elevated feeding rates, to the wide tolerance of Aurelia scyphisto mae (and 250 Mar Ecol Prog Ser 510: 241–253, 2014

As for Aurelia, a considerable propa- gation capability is achieved, for the most part, through StyB and ST. Again, polyps colonize the available substrate, producing many non-motile particles in short periods of time. S. malayensis is also capable of produc- ing motile particles and detaching from the substrate in order to move from one place to another (Adler & Jarms 2009, present study, A. Schiariti pers. obs.). Furthermore, it can also form planuloids from tentacle tips/ pieces as another mode of producing new polyps and facilitating dispersion (Adler & Jarms 2009). Up to now, the factors that favour the production of these motile products have remained unidentified. However, despite such high propagation capabilities, there has only been 1 report of a S. ma - layensis bloom (Xian et al. 2005 — al- though the image presented in the paper does not resemble Sanderia). How can one explain that one species commonly blooms and the other does not? The answer may lie in a phylo - genetic signal (Dawson & Hamner 2009, Hamner & Dawson 2009), as well as in the factors that regulate strobilation rates and ephyra survival. Although we have maintained cul- tures of these species for several years in the 3 laboratories, we have no clues as to why strobilation is commonly observed in Aurelia but is very rare in S. malayensis. Furthermore, Aure- Fig. 7. Effects of temperature and food supply on reproduction rates and polyp lia species produce ephyrae through density (PD, number of polyps in each experimental unit) of Sanderia malayen- sis scyphistomae after 6 wk. Rr: overall reproduction rates; StyB: ‘Sanderia- polidisc strobilation, while Sanderia type’ budding; ST: reproduction from parts of stolons/stalks; MP: motile bud- utilizes monodisc strobilation (e.g. like tissue particles. Data are mean numbers of each variable ± standard Uchida & Sugiura 1978, Lucas 2001). α deviation. Different letters represent significant differences at = 0.05 Thus, if we double the number of Sanderia scyphistomae we will, at me du sae) to environmental para meters (Lucas 2001, most, double the number of medusae, but, in the case Lucas et al. 2012), it is not surprising that the species/ of Aurelia, if we double the number of scyphistomae lineages of Aurelia are cosmopolitan and exhibit the we might increase the number of medusae 10- to most frequent bloom events. 20-fold. S. malayensis presents, in general, a reproduction In contrast to Aurelia, POD formation has not been strategy similar to that of Aurelia strains, with the ex- observed for S. malayensis (Uchida & Sugiura 1978, ception of PODs. This species exhibits the highest re- Adler & Jarms 2009, present study). Moreover, this production rates ever reported for a scyphistoma species is sensitive to starvation, with very low or under laboratory conditions (Adler & Jarms 2009, G. negligible reproduction rates when food is not avail- Jarms, A. C. Morandini & A. Schiariti unpubl. data). able. Therefore, reproduction rates might be slow Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 251

during low-food periods, and, lacking cysts, the pop- As usual some exceptions can be found, e.g. the ulation might decrease in size. However, this is only sea nettle C. quinquecirrha encysts when tempera- speculation, and further experiments should be per- tures fall below 2 to 4ºC (Cargo & Schultz 1967). Nev- formed to explain which factors control the density of ertheless, differences in experimental conditions Sanderia scyphistomae, their strobilation rates and among studies and other possible factors may have the survival and growth of medusa stages. led to controversial results. Other factors have been PODs can survive extreme conditions, providing reported to induce POD production in addition to protection against chemical attacks (Blanquet 1972), energy supply. Bacterial fouling has been mentioned starvation periods (>3 yr, A. aurita, Thein et al. 2012; as one of the stressful conditions triggering encyst- 6yr, Nemopilema nomurai, Kawahara et al. 2013) ment (Cargo & Schultz 1966, Blanquet 1972). Schia - and predation (Cargo & Schultz 1966, Hernroth & riti et al. (2008) attributed the high POD production Gröndahl 1985). Therefore, it is likely that the capa- rate observed for L. lucerna to a combination of con- bility to encyst has evolved in order to withstand con- spicuous bacterial fouling in culture dishes and the ditions in which polyps would not otherwise survive experimental temperature and feeding conditions. (Arai 2009). The earliest literature speculated that However, observations of several POD-producers PODs were produced under stress to provide protec- (e.g. L. lucerna, C. lactea, C. achylos, C. colorata, C. tion against adverse conditions. Below some food fuscescens) from our laboratory cultures indicate that threshold, A. aurita scyphistomae may shift the allo- unattached scyphistomae are not capable of produc- cation of nutrition from bud formation to POD pro- ing PODs. As described above, the production of duction (Thein et al. 2012). However, contrasting PODs requires the formation and attachment of a results have been provided for other species where stolon. Hence, special substrate properties, and prob- encystment only occurs when food is available and ably the presence of bacterial fouling, influence the POD production is accelerated by enhanced food attachment rate of stolons (Schmahl 1985). There- supply. Interestingly, the species which require food fore, differences in the substrate conditions utilized to produce PODs are those we have classified in this in the different experiments may have influenced work as the POD-producers adopting a mono-mode podocyst production rates through their influence on strategy. The production of PODs increases with food stolon attachment. supply for the rhizostomes L. lucerna, R. pulmo, R. esculentum, Nemopilema nomurai and Stomolophus meleagris (Jiang et al. 1993, Lu et al. 1997, Schiariti CONCLUSIONS et al. 2008, González Valdovinos 2010, Purcell et al. 2012, Kawahara et al. 2013, present study) and for Our results confirm some general trends discussed the semaeostomes Chrysaora pacifica, C. quinque - in the literature, but here we have provided experi- cirrha and Cyanea nozakii (Littleford 1939, Thein et mental data on several scyphozoan groups to corro - al. 2013). Therefore, food availability appears to be a borate speculation in the literature. The main results key factor triggering POD production. While in the of our experiments show that with increasing temper- multi-mode species, starvation and food scarcity pro- ature and food supply the rate of asexual reproduc- mote encystment, in the mono-mode POD-producers, tion of several species is increased. These conditions food is required for reproduction, as it is for budding are major symptoms of the deterioration in marine in multi-mode species and FSB-producers. environments caused by several anthropogenic actions The utilization of cysts to withstand adverse condi- that favour jellyfish (viz. Purcell et al. 2007, Richard- tions or as a reproductive mode appears to be related son et al. 2009). Our data clearly show that at higher to the capability of the species to form polyps by temperatures and levels of available food polyp den- another different and faster alternative, like LB. In sity increases. If polyp density increases, the magni- contrast to the multi-mode Aurelia, POD production tude of blooms is augmented considerably. However, by mono-mode species requires a certain level of the potential for a species to regulate its polyp den- energy supply. Above this threshold, POD produc- sity depends on the capability of adapting its repro- tion rates increase with temperature and food avail- ductive strategy in response to the environmental ability, as observed in our present work and in previ- conditions. ous studies (e.g. Thein et al. 2013). Even in Aurelia, Beyond the recruitment rate of settled planulae, once encystment has been triggered, POD produc- polyp density is exclusively dependent on asexual tion rates appear to increase with temperature (Thein reproduction rates. Thus, those species capable of et al. 2012, present study). producing non-motile particles (LB, LBst, ST) would 252 Mar Ecol Prog Ser 510: 241–253, 2014

exhibit the highest reproduction rates, followed by quinquecirrha. Biol Bull 142:1−10 those producing FSBs and PODs. In the latter case, Cargo DG, Schultz LP (1966) Notes on the biology of the sea nettle, Chrysaora quinquecirrha, in Chesapeake Bay. not only the slowest reproduction rates, but also the Chesap Sci 7:95−100 need for 2 sets of environmental stimuli in order to Cargo DG, Schultz LP (1967) Further observations on the produce new polyps (one for encystment, another for biology of the sea nettle, Chrysaora quinquecirrha, and excystment), makes this mode of reproduction the in Chesapeake Bay. Chesap Sci 8:209−220 CIESM (The Mediterranean Science Commission) (2001) slowest of those observed to be utilized for propa - Gelatinous zooplankton outbreaks: theory and practice. gation. CIESM Workshop Series 14, CIESM, Monaco On the other hand, species capable of producing Dawson MN, Hamner WM (2009) A character-based ana - podocysts can reduce mortality rates through encyst- lysis of the evolution of jellyfish blooms: adaptation and ment. Although it is true that this also reduces actual exaptation. Hydrobiologia 616:193−215 Fitt WK, Costley K (1998) The role of temperature in survival polyp density, these populations have the potential to of the polyp stage of the tropical rhizostome jellyfish re-grow once favourable conditions have returned, Cassio pea xamachana. J Exp Mar Biol Ecol 222:79−91 enhancing their blooming potential as described by Fuentes V, Straehler-Pohl I, Atienza D, Franco I and others Kawahara et al. (2013). Alternatively, the species that (2011) Life cycle of the jellyfish Rhizostoma pulmo (Scy - phozoa: Rhizostomeae) and its distribution, seasonality produce motile particles can potentially avoid or and inter-annual variability along the Catalan coast and reduce competition for space and food by reaching the Mar Menor (Spain, NW Mediterranean). Mar Biol new settlement sites. At the same time, we observed 158:2247−2266 that the FSB-producers are among this group, which, Gohar HAF, Eisawy AM (1960) The development of Cassio- as mono-mode species, have no other alternatives for pea andromeda (Scyphomedusae). Publ Mar Biol Sta Ghardaqa (Red Sea) 11:147−190 propagation. These species are mostly tropical and González Valdovinos MI (2010) Alimentación y desarrollo more sensitive to cold temperatures. Therefore, en la fase pólipo de la medusa bola de cañón Stomo- scyphistomae that are not capable of encystment are lophus meleagris L. Agassiz, 1860 (Scyphozoa, Rhizosto- found throughout the mid- to late summer but are mida). MSc dissertation, Universidad Autónoma de Baja California Sur, La Paz completely absent during winter and early spring (C. Gröndahl F, Hernroth L (1987) Release and growth of Cya - xamachana; Fitt & Costley 1998). In these cases, not nea capillata (L.) ephyrae in the Gullmar Fjord, western only scyphistoma density but also the continuity of Sweden. J Exp Mar Biol Ecol 106:91−101 the species strongly depends on the success of sexual Hamner PP, Dawson MN (2009) A review and synthesis on the systematics and evolution of jellyfish blooms: advan- reproduction and its pelagic stages. tageous aggregations and adaptive assemblages. Hydro- Therefore, the ability of multi-mode species to biologia 616:161−191 switch their reproductive strategies in response to Han CH, Uye SI (2010) Combined effects of food supply and environmental conditions increases the fitness of temperature on asexual reproduction and somatic these species and enhances their potential to bloom. growth of polyps of the common jellyfish Aurelia aurita s.l. Plankton Benthos Res 5:98−105 Hernroth L, Gröndahl F (1985) On the biology of Aurelia Acknowledgements. We thank Valentina Melica for the aurita (L.). 3. Predation by Coryphella verrucosa (Gas- photograph in Fig. 1J and Jack Costello for his helpful com- tropoda, Opistobranchia). A major factor regulating the ments. This work was funded by INIDEP and Grants CON- development of Aurelia populations in the Gullmar ICET PIP 2013 00615 and PICT 2013-1713 (Argentina), and Fjords, western Sweden. Ophelia 24:37−45 Grants 2010/50174-7, 2011/50242-5 and 2013/05510-4, São Hofmann DK, Neumann R, Henne K (1978) Strobilation, Paulo Research Foundation (FAPESP), CNPq 301039/2013-5 budding and initiation of scyphistoma morphogenesis in and CAPES PROEX (Brasil). This study is a contribution of the rhizostome Cassiopea andromeda (: Scypho- NP-BioMar, USP. This is INIDEP Contribution No. 1859. zoa). Mar Biol 47:161−176 Holst S (2012) Morphology and development of benthic and pelagic life stages of North Sea jellyfish (Scyphozoa, LITERATURE CITED Cnidaria) with special emphasis on the identification of ephyra stages. Mar Biol 159:2707−2722 Adler L, Jarms G (2009) New insights into reproductive Holst S, Sötje I, Tiemann H, Jarms G (2007) Life cycle of the traits of scyphozoans: special methods of propagation rhizostome jellyfish Rhizostoma octopus (L.) (Scyphozoa, in Sanderia malayensis GOETTE, 1886 (, Rhizostomeae), with studies on cnidocysts and statoliths. ) enable establishing a new classifica- Mar Biol 151:1695−1710 tion of asexual reproduction in the class Scyphozoa. Mar Jiang S, Lu N, Chen J (1993) Effect of temperature, salinity Biol 156:1411−1420 and light on the germination of the podocyst of Rho pi - Arai MN (2009) The potential importance of podocysts to the lema esculenta. Fish Sci 12:1−4 (Japanese with English formation of scyphozoan blooms: a review. Hydrobio - abstract) logia 616:241−246 Kawahara M, Ohtsu K, Uye SI (2013) Bloom or non-bloom in Blanquet RS (1972) Structural and chemical aspects of the the giant jellyfish Nemopilema nomurai (Scyphozoa: podocyst cuticle of the scyphozoan medusa, Chrysaora Rhizostomeae): roles of dormant podocysts. J Plankton Schiariti et al.: Scyphozoan asexual reproduction and blooming potential 253

Res 35:213−217 han C, Gili JM (2012) Temperature effects on asexual Kikinger R (1992) Cotylorhiza tuberculata (Cnidaria: Scy - reproduction rates of scyphozoan species from the north- pho zoa) — Life history of a stationary population. Mar west Mediterranean Sea. Hydrobiologia 690:169−180 Ecol 13:333−362 Richardson AJ, Bakun A, Hays GC, Gibbons MJ (2009) The Lambert FJ (1935) Jellyfish. The difficulties of the study of jellyfish joyride: causes, consequences and management their life history and other problems. Essex Naturalist responses to a more gelatinous future. Trends Ecol Evol 25:70−86 24:312−322 Littleford RA (1939) The life cycle of Dactylometra quinque- Schiariti A, Kawahara M, Uye Si, Mianzan HW (2008) Life cirrha, L. Agassiz in the Chesapeake Bay. Biol Bull cycle of the jellyfish Lychnorhiza lucerna (Scyphozoa: 77:368−381 Rhizostomeae). Mar Biol 156:1−12 Lu N, Jiang L, Chen J (1997) Effect of temperature, salinity Schmahl G (1985) Bacterially induced stolon settlement in and light on the podocyst generation of Rhopilema escu- the scyphopolyp of Aurelia aurita (Cnidaria, Scyphozoa). lenta Kishnouye. Fish Sci 16:3−8 Helgol Mar Res 39:33−42 Lucas CH (2001) Reproduction and life history strategies of Straehler-Pohl I (2009) Die Phylogenie der Rhopaliophora the common jellyfish, Aurelia aurita, in relation to its (Scyphozoa und Cubozoa) und die Paraphylie der ‘Rhi- ambient environment. Hydrobiologia 451:229−246 zostomeae’. PhD thesis, Universität Hamburg Lucas CH, Graham WM, Widmer C (2012) Jellyfish life his- Thein H, Ikeda H, Uye SI (2012) The potential role of tories: role of polyps in forming and maintaining scy pho - podocysts in perpetuation of the common jellyfish Aure- medusa populations. Adv Mar Biol 63:133−196 lia aurita s.l. (Cnidaria: Scyphozoa) in anthropogenically Melica V (2013) Influence of substrate availability on the perturbed coastal waters. Hydrobiologia 690:157−167 population growth and on asexual reproduction strate- Thein H, Ikeda H, Uye SI (2013) Ecophysiological character- gies of Aurelia sp. polyps. MSc dissertation, University of istics of podocysts in Chrysaora pacifica (Goette) and Trieste Cyanea nozakii Kishinouye (Cnidaria: Scyphozoa: Se - Paspaleff BW (1938) Über die Entwicklung von Rhizostoma maeostomeae): effects of environmental factors on their pulmo Agass. Arbeiten aus der Biologischen Meeres - production, dormancy and excystment. J Exp Mar Biol station am Schwarzen Meer in Varna 7:1−25 Ecol 446:151−158 Pitt KA (2000) Life history and settlement preferences of the Uchida T, Sugiura Y (1978) On the polyp of the Scyphome- edible jellyfish Catostylus mosaicus (Scyphozoa: Rhizos- dusa, Sanderia malayensis and its reproduction. J Fac Sci tomeae). Mar Biol 136:269−279 Hokkaido Univ 21:279−287 Purcell JE, Uye S, Lo WT (2007) Anthropogenic causes of jel- Widmer CL (2006) Life cycle of Phacellophora camtschatica lyfish blooms and their direct consequences for humans: (Cnidaria: Scyphozoa). Invertebr Biol 125:83−90 a review. Mar Ecol Prog Ser 350:153−174 Xian W, Kang B, Liu R (2005) Jellyfish blooms in the Yangtze Purcell JE, Atienza D, Fuentes V, Olariaga A, Tilves U, Cola- Estuary. Science 307:41

Submitted: November 20, 2013; Accepted: March 24, 2014 Proofs received from author(s): May 26, 2014